Abstracts
Ear & Hearing - Advanced Issue Abstracts
January 2002 |
- Speech Perception and Spoken
Word Recognition: Past and Present.
Jusczyk and Luce
- Band-pass Specific Contributions
of Multiple Generators to the Auditory 40-Hz Steady State
Potentials.
Pratt et al.
- What Drives Mechanical
Amplification in the Mammalian Cochlea?
Withnell et al.
- Use of Vocalic Information
in the Identification of /S/ and /•/ by Children with
Cochlear Implants.
Summerfield et al.
- Nucleus 24 advanced encoder
conversion study: Performance versus preference.
Skinner MW, Arndt PL, Staller SJ
- Conversion from the SPEAK
to the ACE strategy in children using the Nucleus 24 cochlear
implant system: speech perception and speech production outcomes.
Psarros CE, Plant KL, Lee K, Decker JA, Whitford LA, Cowan
RSC
- Threshold, comfortable
level and impedance changes as a function of electrode-modiolar
distance.
Saunders E, Cohen L, Aschendorff A, Shapiro W, Knight M, Stecker
M, Righter B, Waltzman S, Tykocinski M, Roland T, Laszig R,
Cowan R
- The Nucleus 24 Contour
cochlear implant system: Adult clinical trial results.
Parkinson, AJ, Arcaroli J, Staller SJ, Arndt PL, Cosgriff
A, Ebinger K
- ADRO (adaptive dynamic
range optimization) for cochlear implants: a preliminary study.
James CJ, Blamey PJ, Martin L, Swanson B, Just Yvette, Macfarlane
D
- Surgical technique for
the Nucleus Contour cochlear implant.
Cohen NL, Roland JT, Fishman A
- A model of a nucleus 24
cochlear implant fitting protocol based on the EAP.
Franck KH
- Speech Perception using
maps based on neural response telemetry (NRT) measures.
Seyle K, Brown CJ
- Three-month results with
bilateral cochlear implants.
Tyler RS, Gantz BJ, Rubinstein JT, Wilson BS, Parkinson AJ,
Wolaver A, Preece JP, Witt S, Lowder MW
- The results in patients
implanted with the Nucleus double array cochlear implant:
Pitch discrimination and auditory performance.
Lenarz Th, Tasche C, Cristofoli T, Lesinski-Schiedat A, Wallenberg
Ev, attmer R, Busby PA, Frohne C
|
 |
Speech
Perception and Spoken Word Recognition: Past and Present.
Jusczyk and Luce
This article presents a representative
review of spoken language perception research over the last
50 years, paying particular attention to the historical and
theoretical contexts within which this research was conducted.
The introductory section describes some of the core issues at
the beginnings of speech perception research: invariance and
constancy, and the search for perceptual units. This is followed
by the first principal area, a discussion of early work on the
discrimination and categorization of speech sounds and how this
work and its implicit bottom-up view of perception affected
speech research. Some of the topics include categorical perception,
speech vs. non speech processing, selective adaptation and normalization.
The second main area of the paper describes more recent efforts
to understand the processes and representations that underlie
the recognition of spoken words instead of individual phonemes.
The discussion includes models such as Cohort Theory, TRACE,
Shortlist, NAM and PARSYS, including the phenomena they explain
and the shortcomings of each model. The third principal area
of the paper describes recent research on how infants develop
the capacity to perceive their native language. This includes
a compelling description of the first author’s elegant
and pioneering research in infant speech perception. Although
many important questions remain unanswered, it is clear that
the research reviewed in this article has been influential in
the design of sensory aids and the development of rehabilitation
strategies for the hearing-impaired. As the field of speech
perception develops, its influence on clinical practice and
sensory aid design is likely to increase even further.
|
Band-pass
Specific Contributions of Multiple Generators to the Auditory
40-Hz Steady State Potentials.
Pratt et al.
The 40 Hz steady state potential
(40HzSSP) is an auditory event-related potential that is recorded
similarly to Auditory Middle Latency Evoked Potentials (AMEP),
but with a stimulus rate approximating 40 Hz. This faster stimulus
rate has the effect that the later components of one response
(those components that occur after the first 25 ms) may overlap
and summate with the earlier components of the following response.
The 40HzSSP has a relatively simple spectrum that is amenable
to easy quantification, and has been used in objective audiometry,
in the estimation of auditory thresholds in children, and in
monitoring of consciousness level. However, the relative contributions
of brainstem and cortical neural ensembles to this potential
have not been established. This study aimed to separate the
composite contributions to the 40HzSSP and to compare them with
their counterparts in the AMEP, by differentially augmenting
different parts of the 40HzSSP using selective lowpass filtering.
It was found that the 40HzSSP is generated by distinct generators
in the brainstem and cortex; brainstem and primary cortical
generators mostly contribute the high frequency of the response,
and later cortical contributions dominate the lower frequencies.
|
What
Drives Mechanical Amplification in the Mammalian Cochlea?
Withnell et al.
The mammalian cochlea does not
simply transduce the mechanical vibration of sound into neural
impulses: it also amplifies it. This amplification is a level-dependent
active process that is responsible for enhanced frequency selectivity
and hearing sensitivity, increased dynamic range, and otoacoustic
emissions. For many years, biophysicists have tried to determine
the exact mechanism that underlies mechanical amplification
in the mammalian cochlea. Recent studies have provided important
information to help constrain possible theories. This article
considers the evidence for and against different theories, including
two models based on a hair cell bundle motor (which is the generally
accepted cochlear amplification mechanism for non-mammals) and
three models based on somatic motility by the outer hair cells.
|
Use of Vocalic Information
in the Identification of /S/ and /•/ by Children with Cochlear
Implants.
Summerfield et al.
Listeners with normal hearing
identify the “s” and “sh” sounds in fricative-vowel
words such as “sea” and “she” based not
only on the fricative noise but also on the adjacent vocalic
segments, and they may perceive these sounds categorically even
when the fricative noise is made ambiguous (for example by replacing
it with another sound whose spectrum is intermediate between
“s” and “sh”). This study examined the identification
of such stimuli in children with cochlear implants using the
SPEAK coding strategy as well as in three control groups: postlingually
deafened adults with the same implant system, children with
normal hearing, and adults with normal hearing. This last group
took the test before and after the stimuli were processed according
to an acoustic stimulation of the SPEAK coding strategy. Implanted
subjects did not use vocalic information as successfully as
their normal hearing counterparts, with no significant difference
between implanted children and adults. Simulated cochlear implant
processing reduced the use of vocalic information made by normally
hearing adults. Acoustic analyses showed that the SPEAK strategy
eliminates some formant transition information in ways that
are compatible with the identification errors made by implanted
subjects. Taken together, these results show that some children
with cochlear implants can extract information from a following
vowel to help identify the preceding fricative sound. The upper
limit on this ability may be determined by the cochlear implant’s
signal processing rather than by the auditory experience of
the child.
|
Nucleus
24 advanced encoder conversion study: Performance versus preference.
Skinner MW, Arndt PL, Staller SJ
Fitting a cochlear implant is
a difficult task that is further complicated by two factors:
the intrinsic complexity of the device (as expressed in the
number of parameters and options that can be adjusted), and
the relative paucity of scientific evidence concerning how patients
should be fit. In the absence of said evidence, many fitting
decisions are taken based on patient preference. One basic and
important decision to be taken when fitting the Nucleus 24 device
concerns the stimulation strategy to be used, because the ACE,
SPEAK, and CIS strategies are available. Is the strategy preferred
by a patient the same one that gives him/her optimum speech
perception? Or is patient preference completely unrelated to
optimum speech perception? This study examined data from 62
postlinguistically deafened adults implanted at 21 centers in
the United States and Canada, who were systematically exposed
to the three stimulation strategies mentioned above. A battery
of speech perception tests was administered to each patient
with each strategy, and patients also responded to a questionnaire
giving their strategy preferences for most listening situations,
the percentage of time they used each strategy, and the strategy
they found gave them the best hearing and understanding of speech
in 19 listening situations. Of the 62 subjects were participated,
37 preferred the ACE strategy, 14 preferred the SPEAK strategy,
five preferred the CIS strategy, and six of them did not prefer
a single strategy. One clear, significant, and important result
from the study is that the strategy preferred by a patient generally
resulted in higher scores than the other strategies. Another
important result is that there were numerous exceptions to this
trend (from 31 to 61 percent, depending on the test). This article
describes in detail many other important results, and ends by
providing useful fitting advice based on the present study as
well as on other recent studies.
|
Conversion
from the SPEAK to the ACE strategy in children using the Nucleus
24 cochlear implant system: speech perception and speech production
outcomes.
Psarros CE, Plant KL, Lee K, Decker JA, Whitford LA, Cowan RSC
Although the ACE stimulation
strategy has proved more advantageous than the SPEAK strategy
for some postlingually deafened adult patients, it remains to
be seen whether this result is applicable to children. This
study assessed whether speech perception and speech production
in seven children using the Nucleus 24 would improve with a
change in speech processing strategy from SPEAK to ACE. Speech
perception was measured using an ABA design, where A represents
the SPEAK strategy and B represents ACE. Subjects had been long-term
users of the SPEAK strategy. They used the ACE strategy for
10 weeks, and then they were switched back to the SPEAK strategy
for four additional weeks. Speech perception scores for words
in quiet were significantly higher when the ACE strategy was
used and there was also an improvement in perception of sentences
in noise when subjects switched from the SPEAK strategy to the
ACE strategy. However, for the sentences-in-noise test, scores
obtained in the final session using the SPEAK strategy were
higher than those obtained in the first session using the same
strategy, which suggests that a learning effect may have influenced
outcomes. The superiority of the ACE strategy was not always
apparent immediately after the first strategy change, as some
subjects showed a decrease during the initial two week period
after changing to the ACE strategy. Speech production results
(which were obtained using an AB design) were less clear cut.
In summary, although some children were able to improve their
speech perception by using the ACE strategy, the differences
in overall performance between the two strategies appeared to
be relatively small.
|
Threshold,
comfortable level and impedance changes as a function of electrode-modiolar
distance.
Saunders E, Cohen L, Aschendorff A, Shapiro W, Knight M, Stecker
M, Righter B, Waltzman S, Tykocinski M, Roland T, Laszig R,
Cowan R
One of the potential benefits
of electrode arrays that lie close to the modiolus (like the
Nucleus 24 Contour array) is that the electrical currents necessary
to reach threshold (T) and comfortable (C) levels are reduced.
In turn, the reduced stimulation levels result in longer battery
life. This study measured T and C levels, electrode impedance,
and the distance between modiolus and electrode contact for
several electrodes in several patients using either the standard
Nucleus electrode array (or “banded” array) or the
Contour array. T and C levels, as well as electrode-modiolus
distance, were significantly lower for the Contour than for
the banded array. In a majority of subjects, there were significant
correlations between electrode-modiolus distance and psychophysical
T and C levels. Electrode impedances were higher for the Contour
array than for the banded array, but this difference was accounted
for by the differences in electrode size. Taken together, these
results suggest that the Contour array is indeed closer to the
modiolus than the banded array, and that this reduced distance
results in lower T and C levels.
|
The
Nucleus 24 Contour cochlear implant system: Adult clinical trial
results.
Parkinson, AJ, Arcaroli J, Staller SJ, Arndt PL, Cosgriff A,
Ebinger K
Results obtained by postlingually,
severely-to-profoundly deaf cochlear implant recipients with
the Nucleus Contour array were assessed using different measures,
some of which were compared to those obtained by users of the
standard banded array. Measures included speech perception scores,
threshold (T) and comfortable (C) stimulation levels, pre- and
post-operative acoustic hearing thresholds and results from
a questionnaire filled out by surgeons. Average T and C levels
were considerably lower for users of the Contour array than
for users of the banded array. Speech perception scores improved
significantly after implantation: for example, average CNC word
scores were 3.9% preoperatively and improved to 38.4% as soon
as 3 months after device activation. Surgeons’ comments
regarding ease of insertion of the Contour array were very positive.
Finally, of 85 patients who had measurable hearing preoperatively
at two or more frequencies in the ear to be implanted, 41 (48%)
still had measurable hearing at one or more frequencies and
32 (38%) had measurable hearing at two or more frequencies postoperatively,
in the implanted ear. In other words, any possible trauma caused
by insertion of the Contour array was not significant enough
to eliminate residual hearing.
|
|
ADRO
(adaptive dynamic range optimization) for cochlear implants:
a preliminary study.
James CJ, Blamey PJ, Martin L, Swanson B, Just Yvette, Macfarlane
D
Although cochlear implants are
an excellent tool to improve speech perception by severely to
profoundly deaf patients, it is likely that further improvements
may be obtained with the development of appropriate signal processing
and stimulation strategies. This study reports the development
and testing of ADRO, a pre-processing scheme that continuously
adjusts the gain in each one of the analysis frequency bands
used with the SPEAK and the ACE speech coding strategies. Nine
subjects were tested in the laboratory, with and without ADRO
pre-processing, at different presentation levels. For speech
presented in quiet there was a significant advantage for the
ADRO condition, but no significant differences between the ADRO
and the standard program were observed when speech perception
was assessed under multitalker babble. Take-home experience
was assessed with preference questionnaires, with five out of
nine subjects indicating a strong overall preference and three
subjects indicating a slight preference for ADRO. In conclusion,
ADRO is a viable alternative to fix the channel gain and may
result in increased perceptual benefit for cochlear implantees
under certain conditions.
|
Surgical
technique for the Nucleus Contour cochlear implant.
Cohen NL, Roland JT, Fishman A
Although they use the same internal
electronics, the Nucleus 24 Contour cochlear implant is physically
different from its predecessor, the Nucleus 24. In particular,
the receiver/stimulator is smaller and the electrode array is
precurved so that it tends to be closer to the modiolus once
a stiffening stylet is removed. This is contrast with the previous
generation banded array, which tended to be closer to the outer
wall of the cochlea. These physical differences have important
implications for surgical technique, which are carefully described
in this article, authored by surgeons from a cochlear implant
center with extensive Contour experience.
|
A
model of a nucleus 24 cochlear implant fitting protocol based
on the EAP.
Franck KH
Several recent studies have proposed
the use of Neural Response Telemetry to obtain electrically
evoked whole nerve action potential (EAP) data that may be used
to estimate threshold and comfort levels in cochlear implant
users. This study makes two further contributions. First, an
EAP-based cochlear implant fitting protocol is proposed that
requires only two loudness judgments from the patient and is
done in a live-voice mode. Second, a retrospective split-half
study design was used: the correlations between psychophysical
mapping levels and EAP data were derived from six of the subjects
and this information was used to predict the psychophysical
mapping levels of a different group of six subjects. The predicted
mapping levels were very close to the behaviorally measured
levels, supporting the feasibility of the clinical use of the
proposed method.
|
Speech
Perception using maps based on neural response telemetry (NRT)
measures.
Seyle K, Brown CJ
Even though T and C levels can
be estimated reasonably well using EAP data, it is important
determined whether the differences between these estimates and
traditional, behaviorally obtained estimates are significant
enough to affect speech perception. This study assessed the
speech perception of 10 adult Nucleus cochlear implant users,
using one of three maps. The "Measured Map" was created
using standard programming techniques (i.e., behavioral estimates
of T and C levels); the "+10/-20 Map" was created
using only the EAP data; and the "Combined Map" used
a combination of EAP thresholds and a single behavioral measure
of T and C levels from electrode ten. Although speech perception
scores using the Measured Map were significantly higher than
those obtained using the other maps, the difference was not
large. This suggests that while maps based on EAP data may not
be optimal, they are of sufficient quality to support reasonable
levels of speech understanding. This result is important because
EAP-based maps may prove to be the most useful when very young
children receive a cochlear implant.
|
Three-month
results with bilateral cochlear implants.
Tyler RS, Gantz BJ, Rubinstein JT, Wilson BS, Parkinson AJ,
Wolaver A, Preece JP, Witt S, Lowder MW
Although bilateral amplification
has become the standard of care when fitting hearing aids (with
the exception of unaidable ears), this practice is still extremely
rare in the case of cochlear implant fitting. This may be a
consequence of the high cost of cochlear implantation combined
with the absence of evidence documenting the potential advantages
of bilateral implantation. In turn, the latter is due to the
fact that few patients have received bilateral implants to date,
and even fewer have been extensively studied. This study evaluated
nine postligually deafened adults who received bilateral cochlear
implants. Speech perception was measured in quiet and with noise
coming from either side, and results are reported for each ear
as well as for both ears with noise coming from either side.
Additionally, a measure of localization ability for broadband
noise (presented either to the right or left of center at 45
degrees azimuth) was obtained, for each ear individually and
for both ears together. With the exception of one subject who
was near ceiling with a single implant, there was a clear “head
shadow” effect: when noise was presented on the same side
as the individually-tested implant, the addition of the implant
on the other side resulted in clear increases in speech perception.
Additionally, speech perception gains beyond the head shadow
effect were documented for five of the nine subjects, who showed
increased speech perception in the bilateral condition either
when noise came from the front (like the signal) or from the
side contralateral to the implant. Finally, localization performance
was better in the binaural case than in the monaural case for
six out of the seven subjects who were tested. These results
suggest that binaural implantation can provide advantages in
speech perception and localization.
|
The
results in patients implanted with the Nucleus double array
cochlear implant: Pitch discrimination and auditory performance.
Lenarz Th, Tasche C, Cristofoli T, Lesinski-Schiedat A, Wallenberg
Ev, attmer R, Busby PA, Frohne C
In patients with obliterated
cochleas, only a reduced number of electrodes can be inserted
with standard cochlear implants, which may result in reduced
auditory performance. A device that can be used to address this
problem is the Nucleus Double Array cochlear implant, a device
that has two electrode arrays with 11 and 10 active electrodes
each, instead of the single 22-electrode array found in the
standard Nucleus device. The 11-electrode array is inserted
into the basal turn of the cochlea and the 10-electrode array
is inserted into the second turn through a second cochleostomy.
This prospective study evaluated speech perception and pitch
scaling by six users of the Double Array, who were tested under
three conditions: basal array only, apical array only and both
arrays simultaneously. The ability to discriminate the pitch
associated with different electrodes was better with the basal
array than with the apical array, and single-array speech perception
was better with the basal one than with the apical one. The
simultaneous use of both arrays resulted in better speech perception
than the use of either array alone, suggesting that in spite
of limited pitch discrimination for its ten electrodes, the
addition of the apical array provides important information
for speech recognition.
|
|
